Communications of the ACM
The Strength of Weak Learnability
Machine Learning
Original Contribution: Stacked generalization
Neural Networks
Cryptographic limitations on learning Boolean formulae and finite automata
Journal of the ACM (JACM)
The weighted majority algorithm
Information and Computation
Boosting a weak learning algorithm by majority
Information and Computation
Machine Learning
Boosting classifiers regionally
AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Combining Classifiers with Meta Decision Trees
Machine Learning
Boosting the margin: A new explanation for the effectiveness of voting methods
ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
Learning When to Trust Which Experts
EuroCOLT '97 Proceedings of the Third European Conference on Computational Learning Theory
Improving boosting by exploiting former assumptions
MCD'07 Proceedings of the 3rd ECML/PKDD international conference on Mining complex data
An empirical study of the convergence of regionboost
ICIC'09 Proceedings of the Intelligent computing 5th international conference on Emerging intelligent computing technology and applications
The effect of distance metrics on boosting with dynamic weighting schemes
FSKD'09 Proceedings of the 6th international conference on Fuzzy systems and knowledge discovery - Volume 1
| Hi-index | 0.00 |
Recently, Freund, Mansour and Schapire established that using exponential weighting scheme in combining classifiers reduces the problem of overfitting. Also, Helmbold, Kwek and Pitt that showed in the prediction using a pool of experts framework an instance based weighting scheme improves performance. Motivated by these results, we propose here an instance-based exponential weighting scheme in which the weights of the base classifiers are adjusted according to the test instance x. Here, a competency classifier ci is constructed for each base classifier hi to predict whether the base classifier's guess of x's label can be trusted and adjust the weight of hi accordingly. We show that this instance-based exponential weighting scheme enhances the performance of AdaBoost.